I've been looking for the Majestouch NINJA [BrownSwitch/Fullsize/US ASCII] (1). If I find one, I hope I can get the proposal past my minister of finance.
It's interesting to watch ordinary people attempt to dabble in currency. How many criminal organizations, governments, and financial institutions (redundant?) the world over have people who wake up everyday just to manipulate the value of currency? For example, a common criticism of China is that they purposely manipulate their currency to devalue the US dollar. In which ways could such organizations impact a new currency like bitcoin? Simple theft doesn't work very well unless immediately converted to cash, just like tangible goods being fenced, often below true value. Was it really just theft?
I'm just an armchair economist, but I have read real economist argue that a strong US dollar limits the US's ability to repatriate it's manufacturing base. Also, I don't think China so much increases the dollar as it limits the growth of the Yuan in measure with technological and capital improvements that allow it to keep its export market strong and keep manufacturing business in China.
So how long will it be before someone can take a picture of a person and find all the similar matches on Tinder modulo any profile preferences and desired attributes? Maybe save it for later and use it in a completely different location? In other words, take a picture and then have an app that finds as many matches for that person all while applying any filter preferences the user has in his or her own profile. One person may see exciting matching opportunities. Others may use these modern technology aids in becoming the ultimate modern predator. It's one thing when you want to find a great date. It's quite another when someone uses it to find your son or daughter.
Why add more moving parts when they don't do anything but make more work? Scanning is a helpful idea, but not AV scanning. Regular vulnerability scanning can assess the platform security. At the very least, it can warn about potential security holes. It might also be plagued with false positives, causing more work for no added benefit. Safely running services on the internet is hard.
> Why add more moving parts when they don't do anything but make more work?
Because in this case we're talking about intentionally accepting files from users to either integrate into the system or offer to other users. Why would you not at least check files for cleanness and reject any that fail instead of blindly accepting them because you wanted to enable file uploads.
Vulnerability scanning isn't going to tell you much when you want to accept files from users.
I especially appreciate PostgreSQL's documentation. I find it easy to locate, search, and understand. I certainly have not had that experience with Oracle docs.
Oracle's documentation is excellent and is WAY more comprehensive than PostgreSQL. For example, most of PostgreSQL's SQL functions have a single line in a table to describe them, whereas each one in Oracle has it's own section with a full description, examples, etc.
I agree that the relative small size of PostgreSQL documentation and the fact that it fits in a single "book" compared to the many books that comprise the Oracle documentation can make it easier to find something. But can't say I've ever seen a product with as much quality documentation as Oracle.
As a user, you want to look in "SQL Language Reference" first, followed by "Data Warehousing Guide" (found in a separate folder on the left side). As an administrator, "Concepts" is required reading followed by "Administrator's Guide".
> Oracle's documentation is excellent and is WAY more comprehensive than PostgreSQL.
I'd generally agree that Oracle's documentation is more extensive than PostgreSQL's in many respects, but Oracle's is also substantially less well organized than PostgreSQL's.
That was a bit tongue-in-cheek... Oracle makes an exceptionally solid RDBMS and deserves credit for that. It's just that it hurts every time I write VARCHAR2.
Not only as an example of "single line in a table" but also as an example where a huge volume would not add much value.
I'm struggling at this time to figure out how to produce multiple pages of boilerplate explaining "min()" that would be any more useful than the table.
On the other hand the explanation about nulls really sucks. Its important enough to put in the single line. For those too lazy to click and read, when a noob thinks he wants "SUM(bunchastuff)" what he probably wants is along the lines of using COALESCE or a homemade function that smells like COALESCE. There is a meta discussion that whenever NULLs are possible, and its more than just table definitions for example an overactive WHERE clause, then the NULLs will be a PITA. Defensive SQL coding can be tedious.
I've always found Oracle's database documentation to be reasonably good -- PostgreSQL's is better, but I'd rather be using Oracle's than Microsoft's SQL Server docs.
To be fair, Microsoft's reference material is pretty good - it's standardized, comprehensive and available for every version. You just need to know what to google to find the right page..
The FreeBSD ports system is wonderful because it lets me vary the version of an application without forcing me to replace world. Yum is great when you want a binary distribution that often lags the latest releases by years. What do you do when you need a more recent PostgreSQL than 8.4? Upgrade your OS? snicker Why should I be forced to disrupt my OS when all I want is an application? I develop in the CentOS environment on a daily basis. I long to return to the simplicity and elegance of FreeBSD.
Sure, I'm already using that option. I had to go and find it on my own, at the time. It wasn't as simple as updating the ports collection and installing from ports, but I suppose it's not much different than waiting for a maintainer to update a package and then installing it.
I use ports to mean compile on my own and packages to mean installing precompiled binaries. In my experience, FreeBSD seems more amenable to managing, building, and installing software from source code than a linux distribution. Perhaps I just haven't learned an efficient manner that integrates well with yum/rpm on CentOS? I've always had more dependency nightmares on a linux distro than with FreeBSD. Using FreeBSD's ports collection (single repository) has been less error prone for me than trying to manage all the different yum repos one must manually organize to get up to date software. Am I doing something wrong or is it really that clunky?
Yes, I guess if what you want is to build software from source, then the ports are not THAT bad; but when you want to get a little more generic or care about deployment time (especially when dealing with cloud instances) then the fun goes away. Plus the rolling release nature of the ports sucks arse..
I'm not advocating against FreeBSD, I like FreeBSD and I want to see it again on the table, but the ports are hurting it and its adoption, it needs to go away.
You are quite correct that installing from source doesn't really scale in hosting services. It does take too long and leads to incongruent deployments that may block migration. I don't believe any distribution is well suited for cloud deployment out of the box. I usually deploy from a suite of prepared images via netboot, then customers are free to compile from source or use canned packages.
Transition to a less demanding role, if needed, and look for other areas where you can contribute the result of your experience.
For example, if you have passion for *BSD, some linux distribution, or some other community software project you could contribute. You could start by fixing bugs and submitting patches.
OR, skip out on software development completely. Find a way to get more free time and then study a foreign language or learn how to play a musical instrument. Maybe the issue isn't what you are doing but what you aren't?
BSD vs GPL was already mentioned. Perhaps now we can discuss how DRM only hurts paying customers of protected content? Oh, you bought access to some licensed content? Great! Thank you. As a reward we will now severely limit how you consume the good or service.
1. http://www.diatec.co.jp/en/det.php?prod_c=772